Explore the vast range of titles available.

[...]studies in this field can be divided into two aspects: first, new end-to-end neural network models for merging constituent parts during the image fusion process; Second, the embodiment of artificial neural networks for image processing systems. [...]a feature pyramid module (SC-FP) based on spatial and channel attention can perform the multi-scale fusion of features accompanied by feature selection. Electricity transmission line monitoring in hazy weather will face some problems, such as reduced contrast and chromatic aberration. [...]Zhang M et al. proposed an image defogging algorithm for the electricity transmission line monitoring system. In this research, an optimized quadtree segmentation method for calculating global atmospheric light was proposed. [...]the detail sharpening post-processing based on visibility and air light level was introduced to enhance the detail level of electricity transmission lines in the defogging image.

[...]studies of image fusion can be divided into two areas: first, new end-to-end neural network models for merging constituent parts during the image fusion process; second, the embodiment of artificial neural networks for image fusion systems. In this method, the cluster number K was calculated by fusing local binary patterns and gray-level co-occurrence matrix characteristic values. [...]the clustering and segmentation operation can be performed on Tujia brocade images by adopting a Gaussian mixture model to get a rough preliminary segmentation image. In the fourth paper, Wu et al. propose fractional wavelet-based generative scattering networks (FrScatNets) in which fractional wavelet scattering networks are used as the encoder to extract image features, with deconvolutional neural networks acting as the decoder, to generate an image. [...]the authors also developed a feature-map fusion method to reduce the dimensionality of FrScatNet embeddings. [...]wavelet decomposed multiscale magnitude spectra for every single channel were produced.

To create a safe bicycle infrastructure system, this article develops an intelligent embedded learning system using a combination of deep neural networks. The learning system is used as a case study in the Northumbria region in England’s northeast. It is made up of three components: (a) input data unit, (b) knowledge processing unit, and (c) output unit. It is demonstrated that various infrastructure characteristics influence bikers’ safe interactions, which is used to estimate the riskiest age and gender rider groups. Two accurate prediction models are built, with a male accuracy of 88 per cent and a female accuracy of 95 per cent. The findings concluded that different infrastructures pose varying levels of risk to users of different ages and genders. Certain aspects of the infrastructure are hazardous to all bikers. However, the cyclist’s characteristics determine the level of risk that any infrastructure feature presents. Following validation, the built learning system is interoperable under various scenarios, including current heterogeneous and future semi-autonomous and autonomous transportation systems. The results contribute towards understanding the risk variation of various infrastructure types. The study’s findings will help to improve safety and lead to the construction of a sustainable integrated cycling transportation system.

Word embeddings are a powerful machine-learning framework that represents each English word by a vector. The geometric relationship between these vectors captures meaningful semantic relationships between the corresponding words. In this paper, we develop a framework to demonstrate how the temporal dynamics of the embedding helps to quantify changes in stereotypes and attitudes toward women and ethnic minorities in the 20th and 21st centuries in the United States. We integrate word embeddings trained on 100 y of text data with the US Census to show that changes in the embedding track closely with demographic and occupation shifts over time. The embedding captures societal shifts—e.g., the women’s movement in the 1960s and Asian immigration into the United States—and also illuminates how specific adjectives and occupations became more closely associated with certain populations over time. Our framework for temporal analysis of word embedding opens up a fruitful intersection between machine learning and quantitative social science.

Spiking neural networks (SNNs) are inspired by information processing in biology, where sparse and asynchronous binary signals are communicated and processed in a massively parallel fashion. SNNs on neuromorphic hardware exhibit favorable properties such as low power consumption, fast inference, and event-driven information processing. This makes them interesting candidates for the efficient implementation of deep neural networks, the method of choice for many machine learning tasks. In this review, we address the opportunities that deep spiking networks offer and investigate in detail the challenges associated with training SNNs in a way that makes them competitive with conventional deep learning, but simultaneously allows for efficient mapping to hardware. A wide range of training methods for SNNs is presented, ranging from the conversion of conventional deep networks into SNNs, constrained training before conversion, spiking variants of backpropagation, and biologically motivated variants of STDP. The goal of our review is to define a categorization of SNN training methods, and summarize their advantages and drawbacks. We further discuss relationships between SNNs and binary networks, which are becoming popular for efficient digital hardware implementation. Neuromorphic hardware platforms have great potential to enable deep spiking networks in real-world applications. We compare the suitability of various neuromorphic systems that have been developed over the past years, and investigate potential use cases. Neuromorphic approaches and conventional machine learning should not be considered simply two solutions to the same classes of problems, instead it is possible to identify and exploit their task-specific advantages. Deep SNNs offer great opportunities to work with new types of event-based sensors, exploit temporal codes and local on-chip learning, and we have so far just scratched the surface of realizing these advantages in practical applications.

In recent years, deep neural networks have been successful in both industry and academia, especially for computer vision tasks. The great success of deep learning is mainly due to its scalability to encode large-scale data and to maneuver billions of model parameters. However, it is a challenge to deploy these cumbersome deep models on devices with limited resources, e.g., mobile phones and embedded devices, not only because of the high computational complexity but also the large storage requirements. To this end, a variety of model compression and acceleration techniques have been developed. As a representative type of model compression and acceleration, knowledge distillation effectively learns a small student model from a large teacher model. It has received rapid increasing attention from the community. This paper provides a comprehensive survey of knowledge distillation from the perspectives of knowledge categories, training schemes, teacher–student architecture, distillation algorithms, performance comparison and applications. Furthermore, challenges in knowledge distillation are briefly reviewed and comments on future research are discussed and forwarded.

Toward smart building and smart home, floor as one of our most frequently interactive interfaces can be implemented with embedded sensors to extract abundant sensory information without the video-taken concerns. Yet the previously developed floor sensors are normally of small scale, high implementation cost, large power consumption, and complicated device configuration. Here we show a smart floor monitoring system through the integration of self-powered triboelectric floor mats and deep learning-based data analytics. The floor mats are fabricated with unique “identity” electrode patterns using a low-cost and highly scalable screen printing technique, enabling a parallel connection to reduce the system complexity and the deep-learning computational cost. The stepping position, activity status, and identity information can be determined according to the instant sensory data analytics. This developed smart floor technology can establish the foundation using floor as the functional interface for diverse applications in smart building/home, e.g., intelligent automation, healthcare, and security. Designing efficient and fast monitoring and response systems for smart building/home applications remains a challenge. Here, the authors propose a smart floor monitoring system developed through the integration of self-powered triboelectric sensing mechanism and deep learning data analytics.

In cyber-physical systems governed by nonlinear partial differential equations (PDEs), real-time control is often limited by sparse sensor data and high-dimensional system dynamics. Deep reinforcement learning (DRL) has shown promise for controlling such systems, but training DRL agents directly on full-order simulations is computationally intensive. This paper presents a sensor-driven, non-intrusive reduced-order modeling (NIROM) framework called FAE-CAE-LSTM, which combines convolutional and fully connected autoencoders with a long short-term memory (LSTM) network. The model compresses high-dimensional states into a latent space and captures their temporal evolution. A DRL agent is trained entirely in this reduced space, interacting with the surrogate built from sensor-like spatiotemporal measurements, such as pressure and velocity fields. A CNN-MLP reward estimator provides data-driven feedback without requiring access to governing equations. The method is tested on benchmark systems including Burgers’ equation, the Kuramoto–Sivashinsky equation, and flow past a circular cylinder; accuracy is further validated on flow past a square cylinder. Experimental results show that the proposed approach achieves accurate reconstruction, robust control, and significant computational speedup over traditional simulation-based training. These findings confirm the effectiveness of the FAE-CAE-LSTM surrogate in enabling real-time, sensor-informed, scalable DRL-based control of nonlinear dynamical systems.

Embedded systems technology is undergoing a phase of transformation owing to the novel advancements in computer architecture and the breakthroughs in machine learning applications. The areas of applications of embedded machine learning (EML) include accurate computer vision schemes, reliable speech recognition, innovative healthcare, robotics, and more. However, there exists a critical drawback in the efficient implementation of ML algorithms targeting embedded applications. Machine learning algorithms are generally computationally and memory intensive, making them unsuitable for resource-constrained environments such as embedded and mobile devices. In order to efficiently implement these compute and memory-intensive algorithms within the embedded and mobile computing space, innovative optimization techniques are required at the algorithm and hardware levels. To this end, this survey aims at exploring current research trends within this circumference. First, we present a brief overview of compute intensive machine learning algorithms such as hidden Markov models (HMM), k-nearest neighbors (k-NNs), support vector machines (SVMs), Gaussian mixture models (GMMs), and deep neural networks (DNNs). Furthermore, we consider different optimization techniques currently adopted to squeeze these computational and memory-intensive algorithms within resource-limited embedded and mobile environments. Additionally, we discuss the implementation of these algorithms in microcontroller units, mobile devices, and hardware accelerators. Conclusively, we give a comprehensive overview of key application areas of EML technology, point out key research directions and highlight key take-away lessons for future research exploration in the embedded machine learning domain.